The Spread of Consensual Unions in Europe as a Diffusion Process

1. The Spread of Consensual Unions in Europe as a Diffusion Process Hans-Peter Blossfeld Bamberg University Tiziana Nazio Universitat Pompeu Fabra, Barcelona

2. - Introduction - Cohabitation as a diffusion process - Description of the diffusion process - Analyses of the diffusion process - Results - Conclusions Outline

3. Introduction In the early 1960s cohabitation was rare and stigmatised everywhere in Europe. Today it is widespread (and changed its meaning), but not to the same extent everywhere: Southern Europe: cohabitation is still rare and practised by a minority West Germany, Austria and The Netherlands: accepted as prelude to marriage France, GB, Norway, East Germany: accepted as alternative to marriage (high rate of extramarital births) Denmark, Sweden: status normatively like marriage

4. Cohabitation as a Diffusion Process Research questions: What drives the diffusion of cohabitation? How do different institutional contexts affect the diffusion process? Data: Family and Fertility Surveys (FFS) Method: Individual level-diffusion analysis with event history models Exponential model with time constant and time varying covariates

5. Key Elements in the Diffusion Process 1. Innovation An innovation is any idea, object, or practice that is perceived as new by the members of the social system. In the 1960s and 1970s cohabitation was rare and an innovative behaviour. 2. Communication channels Communication channels are the means by which information is transmitted to or within the social system (e.g., mass media or interpersonal communication). 3. Time Time relates to the relative speed with which the innovation is adopted: changing rate of adoption.

6. Model of a Diffusion Process N(t) Number of Non-Adopters Process Time t

7. Characteristics of theDiffusion of Cohabitation Diffusion of cohabitation is studied: - from a rare and deviant behavior to a widespread and accepted partnership choice - over the life course and across successive birth cohorts of women - the population of potential adopters is therefore not static but dynamic (there are continuously new cohorts entering and leaving the risk set). The population is continuously in flux.

8. Time-related Dimensions of the Diffusion Process of Pre-marital Cohabitation Pre-cohort adoption Age • Diffusion of Cohabitation is a highly time-related process: • Potential adoption is typically confined to a specific window in the life-course • Highly dynamic population of potential adopters over time Outflow from the risk set (Entry into marital/non-marital union) Peer group adoption Inflow into the risk set (“Ready for partnership formation”) Time (continuous flow of birth cohorts)

9. The diffusion process: mechanismsand indicators Pre-cohort adoption Knowledge-awareness: Every new birth cohort faces an increasing proportion of “cohabiters” among previous birth-cohorts (rising incidence). They will then experience cohabitation as less deviant, or stigmatised, and more socially accepted right from the beginning. Peer group adoption Direct social modelling: Confirmation of attitudes and behaviours through direct experiences “vicarious trials” by similar others, who constitute concrete examples. Not only direct interpersonal contacts but also the perception of the behaviour proper to the occupants of their position (vicarious reinforcement, abstract modelling, ‘structural equivalence’).

10. Peer group adoption: number of prior adopters within woman’s own birth cohort at age t total number of women in woman’s own birth cohort birth cohort age Pre-cohort adoption: number of women belonging to older birth cohorts at age t number of prior adopters among older birth cohorts at age t

11. Sweden France Peer group adoption: Pre-cohort adoption: (not computable for Sweden)

12. West Germany East Germany

13. Spain Italy

14. Relative Advantages ofConsensual Unions • What are the relative advantages of cohabitation for young people? • increasing uncertainty of youth labour markets • increasing uncertainty of the phase of transition into adulthood • long-term commitments (marriage) are increasingly problematic • consensual unions are a flexible living arrangement • it offers safer sexual relationships in a long-term partnership • it offers many of the benefits of marriage (including the pooling • of resources, the economies of scale) that living together provides

15. Rising uncertainty of young peoples’ labour markets makes cohabitation increasingly advantageous: Previous cohorts experiences (rising incidence): knowledge, awareness Peer group experiences (vicarious trials): direct modelling Social influence Perceived relative advantage of But young people face also constraints which render some decisions unfeasible or too costly to consider. Limits can depend on institutional settings on economic resources or on expressions of social influence (norms, social pressure) woman’s adoption cohabitation Opportunities & structural constraints Normative - legal contexts Housing markets Educational systems Labour markets Gender roles Fig. 2.3.1 Individuals’ adoption of cohabitation and institutional contexts in the diffusion process.

16. Hypotheses: Influence of Institutional Contexts Normative context (family traditions, national context, importance of religion, local conditions etc.) Educational expansion (duration of educational participation, level of qualification) Affordable housing (home ownership rates, rental market, laws) (Growing uncertainty in) Labour Markets Changes in gender roles (women’s growing economic independence, male-breadwinner ideology)

17. propensity to move from non-adoption to adoption at time t effect of time varying and time constant individual characteristics effect of the intra-population diffusion process on the rate of individual adoption The Statistical Model

18. Women, born 1954-73, observed from 15 to 39 years of age Controlling for: age, and birth cohort (not shown). Same models for marriage. West East Italy Spain France Sweden Germany Germany In education - 0,35 - 0, 1 8 - 1,26 - 0,6 8 - 0,73 - 0,24 Level of education Primary (ref.) Secondary 0, 21 - 0,29 - 0,2 3 Tertiary 0,25 - 0, 18 Being employ ed - 0,42 (0,15) 0,2 1 Empl. experience 0,06 0,06 n.a. Religiosity - 0,15 - 0,1 6 - 1,00 - 0,5 5 n.a. n.a. Parental Divorce 0,31 0, 39 0,4 2 0,78 0,29 0,25 0,3 4 0,26 1,4 2 1,90 0,59 0,1 8 Single vs. Parents Pregnancy + + + + + + + big + big - big Residence at age 15 Region - south Peer group / pre - cohort ------------------------ See figure s ------------------------ Results: Diffusion of Cohabitation

19. Effects of Cumulative Peer Groupand Pre-cohort Adoption

20. Conclusions The spread of cohabitation can be described as a diffusion process. There seems to be no important intergenerational mechanism working at the early stage of the diffusion process. Cohabitation is rather driven by peer models. Amongst institutional factors,the housing market (also for leaving home) is crucial, especially in the Southern countries. Cohabitation in Italy (and Spain) is restricted to specific groups of the population: highly educated women, who have gained residential independence, (and in the case of Italy) live in the North, work, and grew up in big urban centres. Diffusion process seems to be blocked. No autonomous cohort trend remains after introducing diffusion covariates.